Perplexity

Volume 342, Issue 4 · 6 articles from 4 issues

In This Issue

• Fast Embeddings on GPUs
• Perplexity's AI Agents Helped Build a Database. They Weren't Allowed to Run It.
• Perplexity's New Agent Runs Entirely on Your GPU - with One Expensive Catch
• A Local-First Agent for Private and Cost-Effective Knowledge Work
• Introducing Portable Computer for Local-First AI
• Photon: Building a Retrieval and Ranking Engine From Scratch

Fast Embeddings on GPUs (Perplexity, Friday, September 4th, 2026)

Perplexity engineering details the GPU serving infrastructure behind its embedding and ranking models for AI-native search. Read more

Perplexity's AI Agents Helped Build a Database. They Weren't Allowed to Run It. (The New Stack, Wednesday, September 16th, 2026)

The New Stack examines CobbleDB, the Rust key-value store Perplexity built with AI agents but would not let them operate. Read more

Perplexity's New Agent Runs Entirely on Your GPU - with One Expensive Catch (The New Stack, Monday, September 14th, 2026)

The New Stack covers Perplexity's Portable Computer on Windows, a local AI agent needing a 24GB+ RTX GPU. Read more

A Local-First Agent for Private and Cost-Effective Knowledge Work (Perplexity, Tuesday, August 25th, 2026)

Perplexity Research describes co-designing a harness and post-trained model that run entirely on the user's device. Read more

Introducing Portable Computer for Local-First AI (Perplexity, Tuesday, August 25th, 2026)

Perplexity launches Portable Computer, running its agent entirely on NVIDIA DGX Spark with cloud escalation only when needed. Read more

Photon: Building a Retrieval and Ranking Engine From Scratch (Perplexity, Thursday, September 24th, 2026)

Perplexity built Photon, a Rust-based retrieval and ranking engine that returns 95% of search results in under 230ms. Read more

More from Perplexity · All section posts · This week